Natural History Collections as Primary Data in Ecological Research

Wed, 2017-10-18 08:45 -- maphillips

Prologue: Many of us in the ADBC world look for ways to expand the community of users of museum collections data and to increase the ways in which collections data are used. Recently, in Trends in Ecology and Evolution (TrEE), an opinion piece was published by Scott A. Morrison, et al. titled "Equipping the 22nd-Century Historical Ecologist." In this paper, Morrison, et al. envision gathering ecological data in a way that provides future researchers a lens to best understand today’s ecological communities—a proposition that requires a more integrated relationship between museum collections/collectors and ecologists. At iDigBio, we have a working group Integrating Collections and Ecological Research (ICER) that is exploring ways to strengthen the ties between these two groups. ICER reached out to working group member Christina (Chrissy) Alba, Research Associate at the Denver Botanic Gardens (DBG). Chrissy looped in the DBG reading group, which includes both collections and ecological staff, to discuss the paper, and she worked with Database Associate Rick Levy to write a response to the Morrison, et al. opinion piece. We present their response here, and look forward to community feedback.

Contributed By Christina Alba (christina.alba@botanicgardens.org), PhD, Research Associate in Floristics and Richard Levy, MS, Database Associate Research & Conservation Department at Denver Botanic Gardens

Ecosystems worldwide are being altered by climate and land-use change, making it critical to document contemporary biodiversity. Today’s plant and animal communities have intrinsic biological, cultural, and aesthetic value. But more, knowing which species occur where, and under what conditions, is vital to understanding future ecosystems that may greatly differ from what we see today. If we treat present-day communities as a baseline for comparative work, we must ask: What data should we take today so that biodiversity researchers working in the future can use it to understand their communities of interest? Further, in an era of limited funding, who can and should take this data? And where and how should the data be stored to ensure its availability to future generations? A recent article by Morrison et al. (2017) poses these questions, highlighting both challenges and opportunities for those working in natural history collections and ecological research communities. Scientists in the Research & Conservation Department at Denver Botanic Gardens recently tackled these questions while discussing Morrison et al.’s paper. Here we share some of our thoughts.

What data should today’s biodiversity researchers take?
Specimen vouchering as best practice

Researchers studying biodiversity come from many disciplines, from systematics to field ecology. They ask questions at different scales of organization, ranging from delineating and refining taxonomic groups, to documenting species diversity and distributions across entire communities. Our research at the Gardens represents a microcosm of these varied disciplines and approaches, but the crux of our shared mission is to document and conserve biodiversity. As part of this mission, we strive to voucher the species that we encounter in our research. This includes not only our (traditionally vouchered) floristics studies, but also our ecological research, where vouchering is a less common practice. Our philosophy—that floristics and ecological work should complement and bolster each other—echoes Morrison et al.’s (2017) call to voucher specimens across various disciplines concerned with understanding biodiversity, including field ecology.

Vouchering plants, fungi, and animals is the only way to confirm taxonomic identification and thus ensure that botanical and ecological research is verifiable and repeatable (Heberling and Issac 2017). Proper vouchering includes taking a representative physical specimen, as well as its associated biological and meta-data, and storing each so that they can be accessed in perpetuity. This is no small task, and we understand that the time and infrastructure needed to voucher specimens poses challenges to researchers whose main focus in not collection and curation. But the return on such an effort is becoming increasingly clear. Natural history collections housed in museums and herbaria document biodiversity over deep space and time in a way that other disciplines have not managed. Excitement about the unique role that these collections serve in understanding not just species diversity, but also ecology, is growing. Importantly, the massive digitization effort undertaken by iDigBio has placed natural history collections at the forefront of advances in big data. Unforeseen uses for collections have since provided novel insights into ecological questions such as how climate change is affecting species phenology and distributions. As such, we agree with Morrison et al. (2017) that collections serve a crucial role in supporting the work of our successors, and that ecologists must be brought more fully into the world of collections, both as users and as contributors. So how can we best go about this?

Who can and should grow natural history collections to support biodiversity research?
Building collaborations between collections professionals and field researchers

Morrison et al. (2017) correctly state that chronic under-investment in collection and curation has led to patchiness in specimen data used to address ecological questions. To fill these gaps, they suggest that natural resource managers and field scientists build collections, in some cases independently of traditionally accredited institutions. We agree with the authors’ desire to expand collection efforts and understand that the current funding climate requires innovation to achieve this goal. We further agree that managers and field ecologists are in a great position to make collections. They are often stationed in a single location and can thus make longitudinal collections, or they are traveling to remote locations where collections have never been made (Morrison et al. 2017). But a move toward (even partly) decentralizing our natural history collections requires much consideration. Curators in museums and herbaria have spent decades developing best practices for specimen preservation and use, representing hard-won expertise in everything from materials science to the legalities of sharing valuable artifacts among institutions. We therefore think that Morrison et al.’s (2017) recommendation for principal investigators to earmark funding to work with archivists and other collections professionals is preferable. This collaborative approach avoids stoking competition between the two groups for already-scarce funding, and best leverages the institutional infrastructure already in place. For example, if accredited institutions partner with independent investigators to garner funding, money can be put toward alleviating space constraints in the nation’s inveterate natural history institutions (with dedicated curation staff) rather than building new, decentralized infrastructure from scratch. In turn, independent investigators are relieved of the considerable cost of curating their own collections, yet still have access to all of the data they contain. While harnessing expertise from multiple disciplines is powerful, ecologists have been slow to integrate into the world of collections.

Ways to integrate ecology into collections
Limitations and possibilities

It is first useful to consider why a gap in collaboration exists between the collections and ecological research communities. One likely contributor is that these groups are often delving into their research questions from different angles. For example, a collector may be interested in documenting what exists in a particular time and location, while an ecologist may ask why something exists in a time and location. When considering these different approaches, a dichotomy begins to appear; ecologists are question driven, gathering data collected specifically for a hypothesis-driven study design, while collectors are gathering foundational and universal data for questions that have yet to be determined. Regardless of this divide, the differing approaches complement each other in ways that already show promise and have the potential to grow into fruitful endeavors.

The museum is the paramount place to start when studying ecological change over time. Historical patterns that can be deduced from specimens offer a lens into the past that otherwise wouldn’t exist. Now, imagine a reality in which the people making collections long ago were ecologically minded; they not only deposited specimens from their studies into museums, but also included ancillary information about environmental conditions and made sure that these data stayed attached with the specimens. One example of the unforeseen power of such data-rich collection is Gordon Alexander’s work at the University of Colorado in the mid-twentieth century. While studying the make-up and distribution of grasshopper communities in the northern Front Range, Alexander made thousands of collections that were later deposited in the university’s museum. What makes this story so exciting is that Alexander was keen enough to make his collections along a transect of weather stations, tying years of meteorological data to his museum specimens. Access to these specimens and data made it possible for his work to be verified and replicated, enabling researchers to ask new questions about how the communities have changed over time (Nufio et al. 2010).

Not all historical specimens will be tied to rich supplementary datasets. In fact, issues such as sampling biases or a lack of information can be an obstacle for ecologists looking to use museum specimen data for their research. These potential deficiencies are important to consider when addressing how we can improve methods for collecting specimens: First, to help with addressing biases, who can and should work to increase the numbers of collections being made? Second, what information can we collect along with specimens to make them more useful for future studies? Collections staff partnering with ecologists and vice versa is a crucial step toward making more geographically complete and data-rich collections.

Some approaches we are taking at Denver Botanic Gardens

Requesting that ecologists make specimen collections on top of conducting their field research is a big ask. The Gardens is working to aid budding partnerships between ecologists and collectors by creating clear and simple protocols that make it easy for anyone to deposit data-rich collections into a museum (in our case, an herbarium). By creating and distributing workflows, illustrated protocols, and even videos, our goal is to provide accessible and digestible information that will help train people interested in making collections. To this end, we recently created a GitHub repository that provides access to these resources (https://github.com/ricklevy21/protoColl). While this is the first iteration in an evolving process, the goal is to create a place where collection managers can share their protocols and methods for collecting data-rich specimens. This includes traditional data on location, habitat characteristics, and co-occurring species. However, we are also considering innovative data fields, for example to document the goal of the collection as a way for future researchers to understand the context in which a collection was made. While these training efforts are a great step toward welcoming ecologists into the world of collections, incentives beyond centralizing and curating ecologists’ work may be necessary to convince hardworking field researchers to haul specimens back to the museum. The collections community (and ecologists already participating in the collections realm) must better engage with field ecologists to show that questions about phenomena like rapid evolution of invaders or the phenological effects of climate change have already been addressed with natural history collections. Going forward, programs that link ecologists and collectors who share research interests and are geographically close would be a step in the right direction. We already make such links on a smaller scale here at the Gardens. For example, we conducted a baseline floristics inventory of a riparian corridor on Gardens property before implementing vegetation and hydrological restoration. We will continue to voucher specimens found in our long-term monitoring over time to have a verified list of plant community change in response to these treatments. More broadly, a requirement for better collections practices may also trigger an increase in ecologists partnering with museums. Journals have already begun to raise standards for reproducible science by requiring that datasets be published in online repositories such as Dryad. Perhaps a call from journal editors for scientists to voucher taxa included in their research is a needed next step.

We would like to close by thanking Morrison et al. (2017) for their thought-provoking publication and to encourage biodiversity scientists from all backgrounds to integrate vouchering into their research.

References Cited
Heberling JM and Isaac BL. 2017. Herbarium specimens as exaptations: New uses for old collections. American Journal of Botany 104:963-965. DOI 10.3732/ajb.1700125.

Morrison SA, Sillett TS, Funk WC, Ghalambor CK, Rick TC. 2017. Equipping the 22nd-Century Historical Ecologist. Trends in Ecology and Evolution 32:578-588. DOI 10.1016/j.tree.2017.05.006.

Nufio CR, McGuire CR, Bowers MD, Guralnick RP. 2010. Grasshopper community response to climate change: Variation along and elevational gradient. PLoS One 5(9): e12977. https://doi.org/10.1371/journal.pone.0012977